De-Noising the Speckle Noises of the SAR Image Via the Wavelet Soft-Thresholding Method
Yu Li · Journal of Xiamen University · 2002
Wavelet transformation technique is applied to SAR image de-noising based on careful analysis of Donoho wavelet soft-thresholding method. Level-dependent soft-thresholding algorithm is used to compute threshold and wavelet high-frequency coefficients are thresholded. We get estimated wavelet coefficients, and invert wavelet transformation is performed on them, so that we could suppress speckle noise for SAR images. The experimental results show that the wavelet level-dependent soft-thresholding algorithm is more effective for reducing speckle noise, and can be applied to noised image with the background of speckle noise.